Employing GELLAB, a system of state of the art computer/data base techniques, we automatically detect and relate qualitative and quantitative differences among protein moieties produced by differing experimental conditions in sequences of 2 dimensional electrophoretic gels prepared by our collaborators. Systems yielding significant differences include the HL60 cell line, Friend virus infected murine erythroleukemia cells (both of which show differences depending on whether an erythroid or non-erythroid direction is taken by chemical induced differentiation) as well as the differences in red cell surface proteins correlating with partial immune response to Plasmodium knowlesi. A series of differences in protein synthetic rates which serve to distinguish T-cell lines from a B-cell line has also been demonstrated. Additional refinement of the component algorithms is expected to render GELLAB even more effective a tool for cell biological investigation.